the science of justice...the time period for these data are april 2014 through march 2016. overall,...

THE SCIENCE OF JUSTICE

RACE, DEMONSTRATIONS, AND POLICE USE OF FORCE

APRIL 2017

THE SCIENCE OF JUSTICE: RACE, DEMONSTRATIONS, AND POLICE USE OF FORCE

2 policingequity.org

This report was prepared by TCG on behalf and with the guidance of the Center for Policing Equity.



TABLE OF CONTENTS

Overview .............................................................................................................................................. 4

Executive Summary ........................................................................................................................... 5

History of City Involvement in the Demonstration Policy Review ............................................... 7

Analysis of SPD Data ......................................................................................................................... 8

Availability of SPD Data ........................................................................................................................ 8

Use of Force Counts and Racial Disparities ....................................................................................... 8

Use of Force and Weapons Used ........................................................................................................ 11

Demonstrations and Use of Force ...................................................................................................... 12

Specialized Demonstration Use of Force Approach ........................................................................ 12

Policy Review .................................................................................................................................... 14

Saturation .............................................................................................................................................. 15

Accountability ....................................................................................................................................... 15

Clarity ..................................................................................................................................................... 16

Procedural Justice ................................................................................................................................ 18

Improving Internal Data Collection and Management ............................................................... 20

Conclusion ......................................................................................................................................... 21

Review .................................................................................................................................................... 21

Recommendations ................................................................................................................................ 21



OVERVIEW In recent years, law enforcement executives have increasingly been asked to document and share data, so that third parties might evaluate how fairly police officers are doing their jobs. The Seattle Police Department (SPD) especially has sought partnerships with the Center for Policing Equity (CPE) and other external researchers to assess its policing strategies and to lead policing in the nation with respect to civil rights and public accountability.

This report analyzes the SPD’s Crowd Management 14.090—POL (the “Crowd Management Policy”) and all SPD policies governing officers’ responses to demonstrations. The aim of this report is to provide the SPD with a meaningful assessment of its policies with respect to management of large public assemblies (described in this report as “demonstrations”), to identify improvements the department could make in responding to these events.



EXECUTIVE SUMMARY At the request of SPD, CPE was asked to review and evaluate all SPD’s policies with respect to crowd management at demonstrations. CPE was also asked to review use of force data relevant to demonstrations to identify any patterns of force and racial disparities that emerge, particularly as they relate to demonstrations.

The quantitative portion of this report is responsive to SPD’s request that CPE assess the presence of elevated or racially disparate force during their demonstrations. As a result of the desired narrowness of this report, we did not employ the CPE analytic framework as we have in previous reports.1

In these analyses we observe that, while use of force incidents do not spike in the midst of demonstrations, use of more severe, Type II levels of force does. Given the relatively rare occurrence of Type II force, and the relatively higher concentration of it during demonstrations, it is not surprising that community meetings revealed a common perception that protesters are targeted for severe police physical coercion. However, data do not yet support this conclusion. Rather, without information on the reason officers employed force, it is not possible to identify a reasonable set of comparable incidents for comparison. The elevated rates of severe force coupled with the relatively high rates of racial disparity, while not demonstrating bias, are reasons for further community engagement. Future research should focus on identifying factors that contribute to these disparities with an eye towards isolating those factors over which SPD has direct control. CPE also strongly recommends improving data quality to facilitate stronger inferences regarding patterns of police behavior, including greater spatial resolution, and tracking not only where demonstration rallies occur, but also march or parade routes.

By evaluating the clarity, accountability, saturation, and procedural justice of the demonstration policies, CPE assessed their advancement of five key objectives: 1) promoting unbiased policing; 2) improving community relationships; 3) reducing use of force; 4) strengthening oversight and accountability; and 5) improving internal data collection and management.

We found that saturation of SPD’s demonstration policies was excellent: crowd management policies were well integrated throughout SPD’s policy framework. Accountability was also strong: SPD’s innovative disciplinary option of command responsibility is commendable.

CPE also identified areas for improvement with respect to clarity and procedural justice. SPD’s crowd management policy could be strengthened with stronger definitions of what constitutes a crowd management event; who makes that decision; and which circumstances warrant the use of different kinds of resources.

Recommendations:

1. SPD should further pursue explanations of the elevated severe use of force during demonstrations.

2. SPD should collect data on demonstration march routes in order to track police/community interactions more precisely.

3. SPD should work to increase spatial accuracy for their use of force data.

4. SPD should record collective action with riot shields and bicycles as use of force incidents.

1 The CPE analytic framework, first articulated in CPE’s report detailing patterns of racial disparities in use of force and vehicle stops within the Austin Police Department. The framework articulates a conceptual approach for assessing racial discrimination within policing by enumerating three categories of explanations for racial disparities in police behavior: police-level explanations, community-level explanations, and relationship-level explanations. Data collected from SPD does not permit these analyses at this time.



5. SPD should continue to integrate its crowd management policy across other relevant policies contained in the Manual.

6. Command responsibility for use of force at crowd management events should be retained.

7. SPD’s Crowd Management policy should contain definitions of the following: a. what constitutes a crowd management event; b. who may decide whether a crowd management event is occurring; c. under what circumstances a crowd may and may not be ordered to disperse or restricted in

their access to public spaces; and d. with greater specificity, the circumstances under which certain resources or responses may

be warranted.

8. To enhance the transparency and procedural fairness of SPD’s complaints process, we recommend that all complaints procedures be amended to specify the following:

a. the steps involved in investigation of complaints; b. particular decision points at which complainants may be advised of the progress of their

complaint through the investigative process; c. an opportunity for complainants to be advised about how and why a complaint was

resolved the way it was; and d. after disposition of a complaint, a feedback mechanism by which complainants might share

their perceptions of the fairness of the process and outcome of their complaint. This mechanism should solicit recommendations to make the process more just and transparent.



HISTORY OF CITY INVOLVEMENT IN THE DEMONSTRATION POLICY REVIEW

In this section, we describe how the relationship between SPD and CPE began, and why SPD chose to participate in the National Justice Database.

Seattle is renowned for its culture of protests and street demonstrations. Many Seattleites express their views on political affairs—from Black Lives Matter to Standing Rock, from presidential elections to marijuana legalization—by taking to the streets. While these demonstrations may reflect a vibrant culture of civic engagement, large street protests can cost SPD hundreds of thousands of dollars in overtime, and the frequency of such protests poses a unique challenge to policing in the city of Seattle.

May Day protests in recent years have often featured peaceful daytime demonstrations followed by violent anarchist protests in the evening. On May Day 2015, for example, protesters smashed windows, damaged property, and threw rocks, wrenches, and explosive devices at police. Officers used flash balls and pepper spray to control the crowd. Five officers were injured, and sixteen protesters were arrested. In response to these and other high-profile crowd management events, SPD approached CPE for a review of the Department’s crowd management policies. We were approached by Chief O’Toole to evaluate not only policy but the accusations from community groups that the outcome of the department’s crowd management was biased.

This report was delivered in April 2017.



ANALYSIS OF SPD DATA

AVAILABILITY OF SPD DATA2

Are there racial disparities in use

of force incidents?

Is use of force reporting being

evaded?

Are demonstrations quelled with more or

less severe use of force?

Administrative Datasets

Use-of-Force Rich Data Moderate Data Rich Data

USE OF FORCE COUNTS AND RACIAL DISPARITIES

This section examines the reported incidents of force used by SPD officers. We will first present an aggregate overview of use of force incidents, followed by an analysis of incidents by race in order to note any racial disparities. The time period for these data are April 2014 through March 2016. Overall, there was a gradual yet uneven decline in overall reported use of force during this period, as displayed in Figure 1, below.

2 In previous reports, this table would reference a broader range of questions (across the columns) and data fields (down the rows). Because our scope was limited to demonstrations, the fields are limited. Still the table outlines our organizing analytic questions.



FIGURE 1. Quarterly total count of use of force incidents



FIGURE 2. Quarterly use of force rates by race per 100,000 citizens

Figure 2 illustrates the frequency of incidents broken down by racial/ethnic group, from the second quarter of 2014 to the first quarter of 2016. These data show persistent racial disparities, with Blacks subjected to higher rates of force than Whites or Hispanics based on their representation in the population.

FIGURE 3. Aggregate use of force rates by race per 100,000 citizens

Lastly, Figure 3 plots aggregated use of force rates across 2014, 2015, and 2016, and broken down by race.

Analysis of these data compared to the demographics of the area indicate that use of force disproportionately affects Black citizens. These analyses represent only a descriptive approach to



disparities and do not account for crime, poverty, or other relevant factors. They should not be taken to suggest bias per se from SPD officers, as any number of elements may also produce these patterns of force. However, it does provide both a positive impression that use of force is declining (though only slightly), and a sense of scope regarding racial disparity experienced in the past three years by residents of Seattle. Regardless of the reasons for these disparities, it is in the interest of SPD and the people they are sworn to protect to work collaboratively to bring down both raw numbers of use of force incidents and to narrow the racial disparities in use of force.

Our subsequent reports—informed by a wider array of data—will be able to provide more concrete guidance on what factors predict these racial disparities and which of those are within the power of SPD to influence.

USE OF FORCE AND WEAPONS USED

FIGURE 4. Frequencies (rates) of officer use of different weapons by type

Figure 4 shows each of the types of weapons commonly carried and used by Seattle police officers and the rate at which they are applied to Seattle residents, broken down by race and aggregated across 2014, 2015, and 2016. Again, racial disparities are evident, but there are insufficient data with which to form clear conclusions as to the causes of those disparities. Subsequent analyses will provide a clearer picture of the causes of these disparities.



DEMONSTRATIONS AND USE OF FORCE

Generally, during the 2014-2016 period being examined, the highest concentrations of use of force incidents occurred in the heart of the city, where use of force incidents are 20-200 times more likely than on the northern and southern city ends (see Figure 5).3

The ability to understand the spatial distribution of use of force incidents relies on the specificity of location data collected by a police department. For instance, many departments collect precise longitude and latitude data, allowing precise point estimates of where an encounter took place. Other departments might use a single address for an entire university campus, leaving the precise location under-specified in the data. In the case of SPD, there is a high degree of variability in location specificity. For instance, there is fairly precise location data towards the north and south ends of the city. However, there was relatively less well defined location information in the city center.

Generally, the least-defined areas appeared in the data with greatest frequency. Those were Westlake and Seattle Central Community College. This might cause one to conclude that incidents were more tightly clustered at these locations, since in the absence of more precise location data, all locations will be inferred at the geographic center. As a result, incidents in Westlake and Seattle Central Community College all appear to have occurred exclusively at the geographic center of the site due to imputation. Future work by SPD should focus on improving data collection surrounding the routes and movements of demonstrators, in order to clarify the relationship between additional use of force incidents and demonstrations, particularly at these two locations.

As a result of the poor spatial resolution of SPD data, it was not possible to leverage demographic information about Seattle to understand how neighborhood characteristics (such as crime, race, education, housing concentration, etc.) might predict SPD behaviors. Again, better spatial data will permit better understanding of what predicts racial disparities, and how much of that SPD can prevent.

SPECIALIZED DEMONSTRATION USE OF FORCE APPROACH

Analysis of the SPD data surrounding demonstrations shows statistically significant differences in the level of force used during demonstrations, versus that used during normal police duty. Because spatial resolution on use of force data was poor, the below analysis collapses across all of Seattle. The analysis, therefore, examines the average number of use of force incidents during demonstrations and non-demonstration days. Use of force that occurred on the same day as a demonstration, but did not occur at

3 Map Source: Google Maps API

FIGURE 5. Heatmap of frequencies of use of force incidents by location



the same location as the demonstration, was included as a non-demonstration use of force for the purpose of calculating the non-demonstration average.

The analysis indicates that the likelihood of an SPD officer using more severe, Type II force more than doubles during a demonstration, while the likelihood of an officer using minor, Type I force drops by about 85% during demonstrations. In other words, when officers use force at a protest (which is more rare than at non-protests), they use more severe levels of force.

TABLE 1. Average count of use of force incidents and severity of force used during demonstrations, compared to non-demonstration events

Non-Demonstration

Mean Demonstration

Mean Significance

All 5.1107 3.4407 0.0339 Type I - Transient pain, disorientation, or aiming of firearm or beanbag shotgun at a subject 2.0144 0.305 6.00E-04 Type II - Physical injury less than a Type III injury or reasonable expectation to cause physical injury, use of CEW (taser) or OC spray, K9 deployment, or other deterrent tactics 0.7569 1.6271 0.0098 Type III - Officer-involved shooting yielding great bodily harm, deadly force, or loss of consciousness 0.0156 0 0.5686 Type III - Other action yielding great bodily harm, deadly force, loss of consciousness 0.0385 0 0.3435

Other/NA 2.2852 1.5085 0.1947

This suggests that those who attend protests are more likely to witness severe displays of force than they might on an average day in Seattle. This could reasonably result in resident concerns that police use excessive force during protests and/or target protesters. However, simply because residents may witness a rare event more frequently during demonstrations does not mean that SPD uses excessive force or is targeting a particular group. Without information on the degree of resistance or the severity of the crime alleged, it is not possible to determine an appropriate benchmark for use of force. Similarly, without being able to assess the demographics of demonstrators, racial disparities are non-diagnostic of police bias.

In addition to these missing data fields, the data are plagued by other problems. Specifically, the analysis for this report hinged on two data sets, both shared with CPE, but sourced separately by SPD. We compared the two data sets and discovered that one contained 108 incidents missing from the other. In the above table, we opted to use only one of those two datasets which, in our experience, contained stronger, more consistent data. Had we exclusively used the other data set, our analysis would have been altered significantly: incidents of Type II use of force remained statistically significantly increased during incidents of demonstration, but little else could have been deduced.

Finally, there are no mechanisms that require that group behavior (e.g., the use of riot shields or bicycles in skirmish lines designed to physically move protestors) are reported as use of force unless there is evidence of severe bodily injury. This means that SPD officers may engage in collective behavior that applies coercive force to residents that is not recorded. This may result in both the under-reporting of minor force and the over-reporting of major force incidents (as there is greater attention to severe bodily injury when recording police behavior at demonstrations).

Taken together, this analysis provides an explanation for why residents may perceive higher rates of force during demonstrations. It also shines a bright light on areas where improved data capture can facilitate a better understanding of SPD’s attempts at democratic policing.



POLICY REVIEW The primary document source reviewed in this analysis was the Seattle Police Department Manual (“Manual”), updated September 1, 2015. All provisions relating to the policing of demonstrations, including provisions with respect to complaint and discipline processes, were reviewed in this analysis. In November 2015 and January 2016, the CPE’s President and Director of Law Enforcement Engagement also conducted focus groups with line officers, supervisors, members of the monitoring team, the office of Professional Accountability, and community stakeholders. Through these focus groups, CPE sought to gain a sense of how policies were understood and implemented in practice. Additionally, CPE conducted evaluation of prior events documented on video, news clips, and attended a crowd management preparation briefing.

CPE’s usual methodology for policy analysis entails a systematic assessment of policies and procedures related to five key objectives:

1. Promoting unbiased policing; 2. Improving community relationships; 3. Reducing use of force; 4. Strengthening oversight and accountability; and 5. Improving internal data collection and management.

Because CPE’s review was limited to demonstration policies for the purposes of this report, we did not conduct the standard NJD analyses that exploit cross-context comparisons to evaluate the degree to which a PD’s values are fully reflected in their policies. As a result, CPE’s Director of Law Enforcement Engagement and a Senior Research Fellow completed the analyses (as opposed to the policy shop in UC Berkeley). The review of SPD’s Manual identified ten specific SPD policies or provisions relevant to its crowd management practices. The most central of these is Policy 14.090, “Crowd Management” (the “Crowd Management policy”). In this report, all policies relevant to crowd management are referred to collectively as the “demonstration policies.” The specific provisions comprising the demonstration policies are set out in Table 2 below.

TABLE 2. SPD policies relevant to policing during demonstrations

Policy No. Title 5.002 Responsibilities of Employees Concerning Complaints of Possible Misconduct 5.140 Bias-Free Policing

8.500-POL-6 Reviewing Use of Force: Crowd Management 14.010 After-Action Reports 14.080 Task Force Mobilization 14.090 Crowd Management 16.100 Patrol Training and Publications 16.140 Traffic Direction and Control 16.180 Patrol Operations Order 16.230 Issuing Tickets and Traffic Content Reports

The entirety of Title 8, which governs use of force practices in general, is also relevant to crowd management.

The key relevant documents related to demonstration management policy were POL-14.090, “Crowd Management,” 14.080 “Task Force Mobilization”, and 8.500-POL-6: “Reviewing Use of Force: Crowd Management.” The nature, frequency and content of SPD training with respect to crowd management could not be ascertained from the documents provided to CPE. Similarly, none of the documentation



received by CPE addresses whether or how officers are evaluated on their content knowledge after receiving such training.

As mentioned above, CPE’s framework evaluates the saturation, accountability, clarity, and procedural justice of SPD’s crowd management policies across various policy domains and components set out in the Manual.

SATURATION

We assess the saturation of a policy objective by asking: Do numerous policy components work toward a common objective across multiple policy domains? Our review of the SPD demonstration policies found that saturation was excellent.

CPE analysts found widespread cross-policy emphasis on the principle that significant force should not be used against peaceful, nonviolent protesters or those using only nonviolent resistance. For example, the Crowd Management policy, POL 14.090, prohibits use of chemical agents or less-lethal munitions to overcome passive resistance by nonviolent or peaceful protesters. This rule is supported by cross-references in other departmental policies with respect to:

• beanbag shotguns;4 • canine deployments;5 • oleoresin capsicum (OC) spray;6 • use of force review;7 and • labor-management disputes.8

Furthermore, focus groups conducted by CPE researchers indicated that line officers were aware of the demonstration policies, and understood them. Focus groups with community members also found broad awareness among the community that officers were consistently reminded of departmental policies regarding demonstrations. Department personnel at different levels of the organization were consistent in their interpretation of existing policy. Public demonstration policy has been widely communicated both internally to the Department and externally to the public as demonstrated by community knowledge. Discrepancy occurs in the community members’ minds in the application of force and direction to protests involving individuals of color.

ACCOUNTABILITY

We assess accountability by asking: How prescriptive is this policy? How is compliance assured? How do SPD’s policies respond to deviations from this policy? Our review of the SPD demonstration policies found that accountability in SPD written policies regulating demonstrations was strong.

48.300-POL-1#10,prohibitinguseofbeanbagroundsagainstpersonsinacrowdexceptwhereasupervisorapprovesanditispossibletotargetaspecificindividualwhoposesanimmediatethreatofcausingimminentphysicalharm,withreasonableassurancethatindividualsinthecrowdwhoposenoriskofviolencewillnotbehit.58.300-POL-2#7,prohibitinguseofcaninesforcrowdcontrolatpeacefuldemonstrations,andpermittinguseofcaninesonlywithsupervisoryapprovalandafteranordertodispersehasbeenissued.68.300-POL-6,referringdirectlytoCrowdManagementpolicyPOL14.090#10withrespecttouseofOCsprayincrowds.78.500-POL-6,referringdirectlytoPOL14.090andestablishingprotocolsforreview.816.190-POL#3,referringdirectlytoPOL14.090withrespecttouseofforceinpicketingsituations.



At a scene with multiple police officers responding to a shifting crowd, complainants may have difficulty identifying which police officer engaged in a behavior (e.g., use of force) that might give rise to a complaint. SPD offers an innovative response to the accountability challenge posed by this reality: command responsibility for all uses of force in crowd-management events.

Under the Crowd Management policy, an Incident Commander, who must be present at the scene of the demonstration, “retains ultimate responsibility for the decisions of subordinates.”9 The Crowd Management policy requires that, “where feasible,” the Incident Commander or his/her designee be the first to issue an order to disperse.

In addition to reviewing Use of Force reports submitted by officers involved in the crowd-management situation, any Incident Commander who directs or authorizes the use of force in a crowd-management setting must complete a Blue Team entry corresponding to the highest level of force used by any officer during the directed use of force.10

In addition to the ordinary mechanisms for review of individual officers’ use of force, 8.500-POL-6 also creates a specialized Force Review Board for demonstration response. This specialized Force Review Board must include two additional members: lieutenants or captains, selected by the Assistant Chief of the Compliance & Professional Standards Bureau for their “experience with crowd management.” The Board’s review is to “focus on command of the incident rather than the individual officer’s application of force.”11 In particular, the Force Review Board is to determine:

• Whether decisions made concerning crowd management were consistent with policy and Department core principles;

• Whether the incident commander or scene commanders adequately documented their reasons for directing the use of force;

• Whether there were any tactical, training, equipment or staffing deficiencies; and • Whether any use of less lethal force during a crowd management event was properly authorized

and executed.12

SPD’s commendable emphasis on command responsibility for use of force during crowd management events tends to reinforce the objectives of improving community relationships, reducing use of force, and strengthening oversight and accountability.

It is worth noting as one area of improvement, however, that complaints against police officers are considered in performance evaluation and promotions only if they are substantiated. While this provision is often justified with the argument that unfounded or unsubstantiated complaints against an employee should not be harmful to that employee, it leaves SPD vulnerable to concerns that, if its complaints process is not transparent and legitimate, then its promotions, assignments, and ultimately culture might not be legitimate. We address this concern in recommendations around complaints processes below.

CLARITY

We assess the clarity of departmental policies by asking: How well can an officer, employee, supervisor, manager, administrator or a member of the public understand the rules set out in the Manual? Although the demonstration policies demonstrate excellent saturation and strong accountability, the policy is not as

9POL14.090#8.108.500-POL-6#2.118.500-POL-6#4.128.500-POL-6#4.



clear as it could be. Three aspects of SPD’s demonstration policies could be changed to enhance the clarity of the policy guidance: improved along three dimensions:

Definitions: The introduction to the Crowd Management policy states that it “pertains to the Department’s practices of crowd management.” Although section 1 of that policy appears to use “public assemblies” as a synonym for “crowd management events,” neither the Crowd Management policy nor any other section of the SPD Manual defines when a group of people might or might not constitute an event subject to the Crowd Management policy.

Given that “the right of the people peaceably to assemble” is protected by the United States Constitution, the definition of which gatherings are subject to the demonstration policies may engage fundamental constitutional rights.

Along the same lines, the Crowd Management policy is less than clear as to when an order to disperse might be issued. The Crowd Management policy states that an Incident Commander can issue an order to disperse if s/he determines that the crowd presents “an imminent risk to public safety or that large-scale property destruction appears likely,”13 but does not explicitly state whether a crowd can be ordered to disperse if no such risk is perceived. Clear, public, and well-defined rules might help to increase perceived fairness and reduce the likelihood of constitutional litigation.

Clear and specific definitions might help officers, commanders, and members of the public to know in advance whether a group of people engaged in collective activity is or is not subject to SPD’s demonstration policing policies.

A definitions section might address questions such as the following: What is the minimum number of people that might engage the demonstration policies? Does the Crowd Management policy apply only to political and labor demonstrations, or would it also include social, sporting, religious or other gatherings? Does it apply only to outdoor crowds, or can an indoor gathering also qualify as a crowd management event? In borderline cases, which criteria or values should guide that determination?

Responsibility for definition: Assuming that a future policy provides a definition of a “crowd management event” that engages this policy, who holds responsibility to determine whether a group or gathering does or not fit the definition?

Resources and officers for crowd control: Based on the materials reviewed, it was difficult to discern which responses and options were available for officers to respond to crowd management events, and when they were considered appropriate. Who should be there, and what kinds of force are permissible? Policy 14.090—TSK-1, “Responsibilities of the Incident Commander,” indicates at #3 that the Incident Commander can consider using various specialty units, such as bicycle units, mounted units, officers on foot, intelligence officers, or SWAT officers, but offers little guidance as to the circumstances in which such resources might be appropriately deployed. The deployment of these resources is left to the discretion of the incident commander. However, the use of force is most often reviewed as to the individual officer’s application not the deployment decision made by command level personnel. This may have the unintended effect of escalating crowd concern and frequently crowd action.

As described above, SPD policies restrict the use of certain kinds of force (chemical agents, less-lethal munitions, canines) against nonviolent or peaceful protesters. It is silent with respect to when other uses of force (e.g., handcuffs, orders to disperse) might be appropriate. Greater specificity in this regard might improve clarity of understanding, consistency of application, and perception of fairness of SPD’s demonstration policies.

1314.090-POL#9(a),14.090-TSK-3.



PROCEDURAL JUSTICE

The President’s Task Force on 21st Century Policing recommends that police departments should adopt procedural justice as “the guiding principle for internal and external policies and practices to guide their interactions with the citizens they serve.”14 It recommends that law enforcement agencies “establish a culture of transparency and accountability in order to build public trust and legitimacy … [and to] ensure decision making is understood and in accordance with stated policy.”15 In order to “embrace a culture of transparency,” the Task Force recommends, law enforcement agencies should make all departmental policies available for public review.”16 In particular, the Task Force recommends that “agencies should communicate with citizens … swiftly, openly, and neutrally” about allegations of police misconduct.

In this regard, the complaints policies and procedures set out in the SPD Manual appear to fall short: these policies and procedures specify no official procedure or channel by which a complainant might be advised as to the progress of a complaint or the reasons for a particular disposition. Complainants often receive only a mailed letter with the final disposition of their complaint and are not encouraged to remain engaged with the department past this point.

Complaints policies and procedures are set out in SPD Policy 5.002, “Responsibilities of Employees Concerning Complaints of Misconduct” (for complaints of employee misconduct) and Policy 5.140, “Bias-Free Policing,” (for complaints of bias in policing). Complaints such as bias or excessive force that may arise in crowd management settings are subject to these general procedures.

Many elements of the complaints procedures set out in Policies 5.002 and 5.140 are consistent with procedural justice, and seem well designed to improve community relationships and strengthen oversight and accountability. For example, complaints may be made verbally or in writing, from any source (e.g., the subject of a police incident, a witness, a third party, or an employee of SPD), and they may be made anonymously. SPD employees are required to assist the complainant in accepting the complaint and providing information or referrals as to where, how, or to whom to file a complaint.17 Furthermore, SPD employees are required to take action to prevent the loss of evidence of misconduct, and they are forbidden to retaliate against complainants.18 They are subject to a duty to report misconduct that they observe, including uses of force that a reasonable officer would believe to be misconduct.19 Supervisors, commanders and civilian managers are also subject to obligations to investigate and document complaints of bias or other misconduct.20 Moreover, all employees who observe dangerous or criminal misconduct must intervene to stop it.21 SPD also requires annual reporting of bias complaints and their disposition.22

Nonetheless, the procedures for investigation and disposition of civilian use-of-force complaints, as set out in the SPD Manual, are less than transparent. SPD Policy 5.002, the general complaints procedure, specifies no procedure by which a complainant might be apprised of the progress of investigation of a complaint or of the reasons for its disposition. It refers to a manual maintained by the Office of Professional Accountability, but this is not contained in the Manual. CPE is therefore unable to assess the adequacy of OPA complaints investigation procedures, and these procedures are not transparent to complainants.

SPD Policy 5.140, the bias complaints procedure, sets out somewhat more detail about the procedure for handling bias complaints. Like Policy 5.002, it refers to an OPA manual which has not been shared with

14FinalReportofthePresident’sTaskForceon21stCenturyPolicing(May2015),Recommendation#1.1,atp.11.15Id.Recommendation1.3,atp.12.16Id.ActionItem1.3.1,atp.13.17See,e.g.POL5.002;5.140#5.18See,e.g.POL5.002#3,4,6;5.140#4.19See,e.g.POL5.002,5.140#4.20See,e.g.POL5.002,5.140#4.21POL5.002.22POL5.140#8.



CPE analysts and is not transparent to complainants.23 Unlike the complaints investigation procedure set out in Policy 5.002, the bias complaints investigation procedure involves a preliminary investigation and potential resolution by a police supervisor. The procedures for supervisory investigation are set out at 5.140-PRO-1 “Handling a Bias-Based Policing Allegation.” This procedure specifies no steps by which a complainant might be apprised of the progress of the investigation of his or her complaint, advised of the outcome, or by which a complainant might receive an explanation as to why the complaint was resolved the way it was.

23Afterapreliminaryinvestigationbyasupervisor,biascomplainantsmaybeofferedreferralofthecomplainttoOPAforinvestigation.POL5.140#6,7.



IMPROVING INTERNAL DATA COLLECTION AND MANAGEMENT As noted above, CPE was unable to conduct a quantitative analysis of SPD use of force at demonstrations because no data were shared in response to requests for non-demonstration uses of force. Without a baseline for comparison, data on use of force at demonstrations cannot be analyzed meaningfully. Additionally, certain documents, such as the OPA Manual, training, and curricula might facilitate a more in-depth and informative assessment of the clarity, accountability, saturation and procedural justice of SPD’s demonstration policies.

Furthermore, although increased documentation and sharing of data are important concerns, they rely upon acceptable levels of data quality. SPD must prioritize sanitation and standardization of data collection. Ideal data is machine-readable, automated, and fully identified data. Although SPD has taken great strides to maintain data on certain activities, analysis will be hindered or prevented if improvements are not made to it.

Finally, SPD must address the inherent issues with maintaining two means of collecting information, if neither one is considered authoritative. The analysis for this report hinged on two data sets, both shared with CPE, but sourced separately by SPD. We compared the two data sets and discovered that one contained 108 incidents missing from the other. Although redundant insight could make data collection more exhaustive, SPD must be capable to understand the differences between them and resolve discrepancies. As it stands, our analysis is compromised by uncertainty over the correctness of the underlying data.

If, in the future, this information were to be shared with external reviewers, a quantitative and more thorough qualitative analysis might enable SPD to evaluate whether and how its policing practices conform to and implement the rules set out in the Manual.



CONCLUSION At this stage of the analysis process, it would be premature to assign comparative ratings to departmental policies. Rather, this analysis offers a narrative overview that aims to identify ways in which SPD demonstration policies, as written, appear to advance the five key policing objectives, and to identify areas in which there may be room for additional clarity or policy development.

To the extent that CPE continues to collaborate with SPD and review its policies, procedures and practices with more comprehensive information, it will be possible to more concretely assess how likely SPD’s crowd management policies are to advance these objectives, and to pinpoint specific areas that may warrant further development.

REVIEW CPE reviewed over 750 pages of the SPD Manual, evaluating its demonstration policies for their advancement of five key objectives: 1) Promoting unbiased policing; 2) improving community relationships; 3) reducing use of force; 4) strengthening oversight and accountability; and 5) improving internal data collection and management. Each dimension was evaluated using qualitative and analytical techniques, and assessed for saturation, accountability, clarity, and procedural justice. SPD’s policies were evaluated as excellent with respect to saturation and strong with respect to accountability. SPD’s integration of its crowd management policies throughout its policy environment is to be commended, as is its innovative structure of command responsibility for use of force in crowd management.

Areas for improvement were identified with respect to clarity and procedural justice. SPD’s crowd management policy would be strengthened with stronger definitions of what constitutes a crowd management event, who makes that decision, and which circumstances warrant the use of different kinds of resources.

RECOMMENDATIONS 1. SPD should further pursue explanations of the elevated severe use of force during demonstrations.

2. SPD should collect data on demonstration march routes in order to track police/community interactions more precisely.

3. SPD should work to increase spatial accuracy for their use of force data.

4. SPD should record collective action with riot shields and bicycles as use of force incidents.

5. SPD should continue to integrate its crowd management policy across other relevant policies contained in the Manual.

6. Command responsibility for use of force at crowd management events should be retained.

7. SPD’s Crowd Management policy should contain definitions of the following: a. what constitutes a crowd management event; b. who may decide whether a crowd management event is occurring; c. under what circumstances a crowd may and may not be ordered to disperse or restricted in

their access to public spaces; and



d. with greater specificity, the circumstances under which certain resources or responses may be warranted.

8. To enhance the transparency and procedural fairness of SPD’s complaints process, we recommend that all complaints procedures be amended to specify the following:

a. the steps involved in investigation of complaints; b. particular decision points at which complainants may be advised of the progress of their

complaint through the investigative process; c. an opportunity for complainants to be advised about how and why a complaint was

resolved the way it was; and d. after disposition of a complaint, a feedback mechanism by which complainants might share

their perceptions of the fairness of the process and outcome of their complaint. This mechanism should solicit recommendations to make the process more just and transparent.

Center for Policing Equity

524 West 59th St Room 6.63.14 NY, NY 10019

310.206.3438 Email: [email protected]

@PolicingEquity

Center for Policing Equity

the science of justice...the time period for these data are april 2014 through march 2016. overall,...

Documents